Video Game Spaces
Image, Play, and Structure
in 3D Game Worlds
Michael Nitsche
The MIT Press
Cambridge, Massachusetts
London, England
© 2008 Massachusetts Institute of Technology
All rights reserved. No part of this book may be reproduced in any form by any
electronic or mechanical means (including photocopying, recording, or information
storage and retrieval) without permission in writing from the publisher.
For information about special quantity discounts, please email special_sales@mitpress
.mit.edu
This book was set in Bell Gothic and Garamond 3 by SNP Best-set Typesetter Ltd.,
Hong Kong.
Printed and bound in the United States of America.
Library of Congress Cataloging-in-Publication Data
Nitsche, Michael, 1968–
Video game spaces : image, play, and structure in 3D worlds / Michael Nitsche.
p. cm.
Includes bibliographical references and index.
ISBN 978-0-262-14101-7 (hardcover : alk. paper) 1. Video games. I. Title.
GV1469.3.N58 2008
794.8—dc22
2008019730
10 9 8 7 6 5 4 3 2 1
1
Introduction
Interactive media and their most prominent and most diverse representatives,
video games, have unsettled traditional media theory. They introduced the
meta-medium computer into our living rooms, opening up a range of opportunities so unfamiliar and so diverse that we are still struggling with clear
definitions. They changed the face of established media and media production
and became a billion-dollar industry along the way. This industry, its fan base,
and a range of noncommercial outlets constantly generate new artifacts to
analyze and learn from. Games have developed from a niche market into a
culturally and commercially significant media form. They may still be the
new kid on the block, but they are rapidly growing up.
Games also have entered academia, where they have been investigated
from a variety of perspectives. Occasionally, proponents of different approaches
have engaged in a very lively debate. But we still have to explore much of
the ever-expanding ground covered by the developing media. One challenge
is the disproportion of a still-expanding media form and the often necessarily
slower speed of generating the critical literature about it. Whereas the prophets of cyberspace were restricted to relatively few digital artifacts, often
available to the selected few with access to high-tech research labs, the field
of game studies today faces an overabundance of games to analyze. This body
of work, in many regards, was necessary to support a wider analysis of
new media.
It is incomprehensible that any single theory could do justice to a form as
rich and vivid as the video game. The variety of these games calls for a diversity of analytical approaches: no one approach is sufficient, but many offer
different yet interconnected perspectives. The more this analytical spectrum
grows in width and depth, the richer our picture of the video game becomes.
This book tries to add its share to the spectrum. It will refer to the debates
and key issues in games research but will shine a different light on them from
a very specific angle—that of navigable three-dimensional virtual spaces. From
this perspective, the book will revisit some models that have been examined
before and suggest new ones that have yet to be debated.
3D game spaces allow players to crawl, jump, run, fly, and teleport into
new worlds of unheard-of form and function. The game space we can experience, discover, and manipulate has become endless and at the same time
more accessible than ever. Video game spaces stage our dreams and nightmares
and they seem to get better at it every year. We will tour the landscape
of video games in an effort to discover how the games work, how they are
presented, how they can be read, why they are important, and how they can
be improved.
Following this book’s call for an interdisciplinary approach, its argument
draws on many other disciplines such as literary studies, architecture, and
cinema. Though comparative and interdisciplinary, it does not focus on the
comparison of media as its main goal. Instead, the cross-referencing is the
result of the topic. Spaces in video games can only be discussed through these
references.
Like the 2D desktop metaphors that came before, the growing use of 3D
graphics in video games shows signs of an evolutionary process. New and old
game franchises adapt to the new “standard” or they struggle to survive.
Whether it is Mario’s step into 3D in Super Mario 64 (Miyamoto 1996), the
polygon worlds of the massively multiplayer online title Meridian 59 (Sellers
1996), the 3D spaces in real-time strategy games such as Ground Control
(Walfisz and Andersson 2000), or the jump from a top-down view in Grand
Theft Auto II (Akiah, Conroy, and Hirst 1999) to the navigable 3D world of
Liberty City in Grand Theft Auto III (Filshie et al. 2001), navigable 3D worlds
have become a critical factor of game development. The market is dominated
by 3D graphics in a way that suggests an overall transformation. This also
affects the hardware of game systems. 3D graphics have shaped the hardware
development of personal computers in the form of specialized graphic cards
at least since the mid-1990s with the release of the NV1 and Voodoo
1 cards.
Often, 3D graphics have become a sine qua non in the modern commercial
video game world—for better or worse. Yet the use of 3D graphics for its own
sake cannot be the goal but rather a means to achieve a more complex task:
the generation of fictional worlds in the player’s imagination that grow from
a comprehension of the 3D representations. Like written text or 2D graphics,
Chapter 1
2
3D worlds have unique ways to support this imaginary work. This raises
questions: How do game spaces achieve this effect? What are their key qualities? How can we improve their impact?
To answer, we have to turn to the player and the player’s experience.
Although this book will look at a variety of design issues and structures,
its fundamental principle is to examine games in terms of the player experience. Space is an important element of this experience—and a very challenging one. The argument here is that game spaces evoke narratives because the
player is making sense of them in order to engage with them. Through a
comprehension of signs and interaction with them, the player generates new
meaning. The elements that are implemented in the game world to assist in
the comprehension will be called “evocative narrative elements,” because they
do not contain a story themselves but trigger important parts of the narrative
process in the player. These processes can lead to the generation of a form
of narrative.
Such an approach has obvious parallels to semiotics. Video game worlds
depend on representation and sign systems and the audiovisual presentation
is an important section in this book. But the main argument will always
return to the concept of space and spatial experience. Game spaces are
approached not as foregrounded spectacles based on visual cues such as perspective and parallax but as presented spaces that are assigned an architectural
quality. The discussion will concentrate on their spatial structures and how
players can interact with them. In the games discussed, players want to engage
not with the screen but with a fictional world these images bring to mind.
This indicates the references to phenomenology that guide this investigation.
Experience, comprehension, and spatial practice are phenomenological key
elements that reappear throughout this discussion of virtual space. But how
this experience is generated in game worlds is new. As much as we can learn
from these approaches, there are fundamental differences in the way experience
of space happens in nondigital settings and the necessarily mediated way we
encounter game worlds.
The screen remains an important layer as it is mainly through the screen
that the game worlds can unfold and become accessible to today’s player. Video
game worlds are navigable spaces that offer a wide range of interactions, but
they are also spaces told to us using certain forms of presentation. This mediation is an important factor in the narrative processes connected to game worlds.
It is in these presentations that we can find filters, which are constantly at work
in 3D video games. Consequently, this book is structured in three main sections: one on the reappearing question of narrative and structure in game
worlds, one on presentation, and one on the functionality of game spaces.
Introduction
3
If this book has to be associated with a specific school or set of approaches
to game research, it would be with a player-focused and experience-driven
tier. This is a result of the topic at hand: the question of space. It is only
logical that the reader will find chapters on identity and player positioning
as well as on camera work and architectural theory. At the same time, this is
a comparative survey driven by practical analyses of existent video games.
Therefore, the reader will find references to numerous games and game
worlds.
Ultimately, this book is an exploration of the new universe that opened up
with the introduction of 3D graphics to games. This change has literally added
a new dimension to video games that calls for a reorientation of games research
and design and asks us to adjust our thinking about video games. Due to the
interdisciplinary nature of this undertaking, the book most certainly has to
miss a number of relevant sources and concepts but is intended to provide
balanced arguments and detailed discussion.
1.1 Video, Game, and Space
Space has been a central issue for the study of digital media since the introduction of cyberspace (e.g., Benedikt 1991) and Multi-User Dungeons (MUDs)
(e.g., Anders 1998). It has been a defining element of—at times opposing—
studies in games research. Murray argues that spatiality is one of the core
features of digital media (1997), Aarseth starts his discussion of Cybertexts
with a look at labyrinths (1997), Ryan writes about narrative aspects of virtual
spaces (2001b), and other studies of video games include often substantial
chapters on space (e.g., Wolf 2002). Qvortrup edited a trilogy of books on
different aspects of virtual spaces (2000, 2002; Madsen and Qvortrup 2002),
and more recently von Borries, Walz, and Böttger published a large collection
of short essays about spatiality in games and related issues (2007). Most of
the work in this area has been discussed in numerous articles, theses, and
conference papers. Jenkins’s influential essays on game spaces build up from
the mid-1990s (Fuller and Jenkins 1995) to a more rounded view almost a
decade later ( Jenkins 2005) and are only one example. This short list is by
no means a full overview of the existing literature, but it shows that a study
as presented in this book has to be clearly positioned to operate in relation to
the other published works. This book touches on a range of established topics,
introduces new interpretations of them, and suggests some new principles
based on the element of game spaces. It connects various approaches to digital
environments and, thus, has to clarify the field covered.
Chapter 1
4
To avoid generalizations two premises will be applied. The games
discussed
䊏
䊏
have to be available on consumer hardware;
and have to offer navigable 3D environments.
The restriction to available consumer products ties the study to a de facto
analysis of existing worlds and serves as a reality check that prohibits any
overly enthusiastic prophecies. It avoids the danger of promising unrealistic
future wonders. Yet, where necessary, the argument in this book will refer to
additional examples that are not commonly available. Video game spaces are,
by and large, commercial products and thus limited by production costs and
market considerations. In order to avoid a purely commercial perspective, this
study also points to numerous research projects and art pieces. Notably, most
of these projects do not depend on additional hardware or expensive systems,
but are realized with the same technology that drives commercial video games.
The author realized some of these projects himself during his work at the
University of Cambridge and at the Georgia Institute of Technology. It is
the philosophy of the Digital Media program at the School of Literature,
Communication, and Culture at Georgia Tech to combine theory and analysis
of digital media with practical experimentation. In this spirit, the book points
to those kinds of experiments and combines them with discussions of
commercial games.
The second restriction is that each game discussed has to feature an interactive and navigable 3D virtual world. In practice, this means that players can
navigate through the world and interact with elements that are staged in it.
One consequence of this definition is that these kind of worlds can offer different viewing angles to whatever events are realized in them. Clearly, this is
the case for a very large selection of games but it does not cover all games. A
range of titles have very limited spatial features or restrict the virtual environment to other representation forms. These titles will appear as references to
clarify the differences and to provide a historic frame where necessary, but
they are not at the heart of this study.
At the same time, this book will try to avoid a number of possible misconceptions. Navigable 3D worlds possess specific qualities and favor expressive techniques that differ from other formats. Perhaps we can blame the
blanket term “multimedia” for an uncritical blending of different presentation
formats into “virtual spaces.” MUDs, 2D worlds, interactive cinema, and 3D
video games are too often treated as utilizing basically the same concept of
Introduction
5
“virtual space.” That can be a misleading simplification. There are fundamental differences among a space described in a written text, a cinematic space,
and an interactive navigable virtual world. They offer different forms of
interactive experiences based on their specific qualities. Developing these
differences will be one task ahead.
A different kind of mistake is found in some commercial approaches to
virtual worlds. Here, space in video games serves too often as polygon-rich
spectacle and “eye candy” that remains largely unused in the actual gameplay.
It is time to revise this immature misconception and build new connections
along the way for further debate.
The purpose of the two self-imposed restrictions is to keep the focus on
the element of space in video games. Space—like time—is a principle at the
heart of Western as well as Eastern philosophy. The study of space enters into
numerous disciplines, so that we will be able to draw from disciplines as
diverse as (in alphabetical order) architecture, computer science, film studies,
game studies, media studies, narratology, philosophy, psychology, and many
more. All of these disciplines contribute to the analysis of video game spaces
but they cannot all be covered in one book. While this book can easily be
accused of missing important disciplinary affinities, it will assemble relevant
points to form a discussion of game spaces that can serve as basis for even
more specific future work.
Such an interdisciplinary approach requires an understanding of different
terminologies and a clear deployment of them in the new context. It will be
one task to identify and adjust the different terminologies.
1.2
Structure, Presentation, and Functionality
Lefebvre breaks down social space into spatial practice, which secretes a certain
society from another; representations of space, which is “conceptualized space,
the space of scientists, planners, urbanists, technocratic subdividers and social
engineers” (Lefebvre 1991, 38); and representational space, “space as directly lived
through its associated images and symbols, and hence the space of ‘inhabitants’ and ‘users’ ” (ibid., 39). His division informs the approach of this book
but the resulting structure grew more out of the ongoing analysis than from
an upfront subscription to Lefebvre’s theory.
3D graphics open up means of expression to the computer that add a new
dimension to their representation. The use of 3D navigable space changes the
textual qualities of video games and distances them in some way from other
interactive pieces that present their content in a less spatially defined manner.
The first part of the book, on Structure, will look at how 3D space can reshape
Chapter 1
6
the textual qualities of video games. It first considers what the new textual
qualities of these game worlds are and offers definitions of the key terms
“interaction” and “narrative” as two strongholds of games research and important factors in the way players make sense of the worlds presented to them.
The main focus rests on the idea of understanding space and movement therein
by ways of narrative comprehension. Narrative is seen as a form of understanding of the events a player causes, triggers, and encounters inside a video game
space. From this vantage point, part I reframes key principles of narratology
in the context of video game spaces before discussing quests as space-driven
content structures for these narrative elements. Presentation is understood as
the expressive element of video games. It grows from the necessary interpretation of the processed rules and data into some form the player can read. Presentation in current commercial video games uses predominantly audiovisual
components. Thus, the second part of the book concentrates on the roles of
the moving image and sound in video game spaces. The analysis looks more
specifically into cinematic elements at work in video games with a focus on
the construction and guided comprehension of virtual spaces. Part II will
argue that different forms of presentation create a kind of narrative filter
through which player interaction is framed into specific contexts. Important
factors are the focalization through the narration and the guidance of the
player’s reading.
Functionality refers to the interactive access and underlying rules determining what the player can do in the game space and what the space can do to
adjust that. Part III looks at the elements that shape the development of the
actions as they unfold in 3D worlds. Means of functionality are, for example,
artificial intelligence (AI), complex rule systems, and interface design. These
topics have been discussed in other works to far greater length. The aim here
is to concentrate on the functionality of virtual 3D spaces. This leads to a focus
on architectural principles and the question of a player’s positioning within
the space. A line is drawn from various architectural principles to their actual
use in video games and the player’s interaction with them. Experiential effects
like immersion and presence are discussed to locate the player as active participant in the game world. We arrive at the idea of a social space, a place in
video games. The argument closes with a reference back to narrative, for those
social places are understood as narrative landscapes.
While part I differentiates 3D game spaces from other interactive media,
parts II and III are more focused on the workings of video game spaces as
such. Both are intrinsically different operational features of video games but
are closely intertwined and interdependent. It will be argued that the level of
presentation is a good example for the gradual development of video games
Introduction
7
and has become part of the functionality and accessible to the player. Realizing
the combination of presentation and functionality is a central goal and the
underlying principle of this book. Historically, there seems to be a rift
between the discussion of a game’s audiovisual presentation and its functional
interactive elements. But elements of spatial practice and mediation of space
are always closely interconnected. Space—virtual or real—has to be encountered and interacted with. One has to explore the interaction and the media
that present it. Any concentration on either presentation or functionality but
not both would destroy the holistic principle of spatial experience. Thus, space
forces us to interconnect and build new bridges between two central poles in
game design and analysis.
1.3 What Space?
What is a video game space? The question has to be answered to avoid introducing a nebulous concept to the core of this argument. The difficulties in
describing and limiting the term do not come as a surprise. The spatial is
connected to and debated in literally every possible arena of human thinking.
We cannot reduce such a holistic principle to any single frame or assume a
single concept of “the space.” If space is such an all-embracing and hardto-restrain phenomenon, the task is not to reduce the term itself but to build
useful frameworks to approach the topic. One has to define the frame for the
specific argument at hand and clarify its context. What are these spaces?
1.3.1 Simulating Theme Parks
Video game worlds are processed. They depend on algorithms and mathematical rules. These rules and processes follow certain models, which makes all
game spaces simulations. Creating these worlds from rules is a form of simulation that follows a preconceived model. But that statement does not really
help us to understand video game spaces as the player experiences them.
Thanks to the freedom of the mathematical models, the resulting worlds
can be anything anywhere, which makes them remarkably indifferent and
somewhat meaningless to start with. There is little given context or relevance
present at the origin of these worlds, neither a history nor a physical constraint
is given to the space itself. That is no problem from a mathematical vantage
point but it makes the resulting virtual spaces almost inaccessible for players.
One has to be a trained expert to deal with them. In order to project the
necessary context into the game worlds and in order to make sense of them,
the player depends on a legible presentation and meaningful functionality.
Players can overcome the nonspecific origins of these spaces with the help
Chapter 1
8
of a designer, who makes the game worlds accessible through interactive
options and shapes their mediation in specific ways. The simulation of an
atom bomb explosion in the Los Alamos National Laboratory and the explosion of the “anomalous materials” in Half-Life (The Half-Life Team 1998)
have very different aims and work very differently as simulations. The same
rings true for more “realistic” sim games like Sid Meier’s work or detailed
driving games.
Yet, the same research labs that developed the atomic bombs famously
provided the Petri dish for modern video games. The Brookhaven National
Laboratory is part of the same Advanced Simulation and Computing program
as, for example, the Los Alamos simulation unit. Both labs were founded to
explore pressing research questions in the “hard” sciences. Nevertheless, about
half a century ago, Brookhaven National Laboratory was the birthplace of the
first documented video game: Tennis for Two (Higinbotham 1958), a prelude
to Pong. The relationship between the armed forces and the game industry in
the United States has been continuous ever since, inviting the term “militaryentertainment complex” (Lenoir 2003). But tracing games only back to war
simulations would be too simplistic.
Unlike the data-driven scientific simulation models, which aim to deliver
reliable findings, accurate training conditions, and highly elaborate answers
to very specific scientific problems, games are centered on the dramatic experience. They do not provide new knowledge through the execution of their code
but instead present engaging questions. The necessary dramatization leads
to a different kind of product that concentrates not on the data, but on the
player.
America’s Army: Operations (Capps 2002) is a first-person shooter game
developed for the U.S. Army and based on Epic’s Unreal engine. It was
intended as a publicity and possible recruitment tool and seems to simulate
an army boot camp, training missions, and real combat situations. Its rule
system might copy military concepts but the way these rules get compromised
through the spectacle of the game illustrates more the tendency of the military
complex to embrace the game community than vice versa. In order to raise
its appeal, America’s Army had to simplify and streamline the education of a
soldier, emphasize the dramatic combat scenarios, and give up the total
control over such training. It remains impossible to track the development of
loyalty, duty, respect, selfless service, honor, integrity, and personal courage
in a world with free chat options and the default “fire at will” setting. In
addition, hacks for avoiding the game’s initial and more rigidly controlled
training phase were available shortly after its release as players started to
utilize their new toy and tweaked it to meet their demands. As a simulation
Introduction
9
America’s Army is of value mainly for tracking social behavior of gamers, not
for military training purposes.
That is why America’s Army serves as a political platform for Joseph DeLappe
in his dead-in-iraq project, which he calls “a fleeting, online memorial”
(DeLappe 2006). In this memorial, DeLappe logs into America’s Army game
sessions as a player named “dead-in-iraq” and uses the chat function to type
in the names of U.S. soldiers who have died in Iraq. The underlying rule
system remains active but its use is subverted. In the case of America’s Army,
such subversion is relatively easy because its title makes the game designer’s
intention obvious. At the same time, the America’s Army website also sells
game-related action figures of decorated war heroes that represent a very different and less critical memorial of military operations. Political activism
obviously works both ways.
But even if the game is based on more complex simulation models, the
author of the game simulation still directs the simulation (Frasca 2003) and
this influence cannot be free of an interpretative stand that includes ideological
patterns (Bogost 2006). These differences between scientific simulations and
games are not only on the level of the creator and code; they also reside on
the side of the player. Players engage with video games not like scientists that
operate a simulation program. They often suspend disbelief when activating
a game and they usually lack an analytical distance to the data. Neither
behavior is allowed in a scientific community. But what games provide is a
reposition of the player into the “active” spot.
Through the loophole of virtual space we are able to reclaim the space of
the action where we do not “look at” but actively visit the center of the action.
When the player’s game avatar Gordon Freeman pushes the “anomalous material” into the particle beam deep inside the Black Mesa Research Facility in
Half-Life and causes a chain reaction, he remains at the heart of the disaster
and has to handle it. The player has to handle it, therefore, because the player
handles Gordon. We are not analyzing the events as neutral onlookers but
share the space with them and have to find our way out of the exploding
chamber and the resulting mayhem. The game might stage the player in the
role of a research scientist, like Gordon Freeman who is introduced as a virtual
MIT graduate, but involvement with the unfolding events is not a scientific
research project. It is a fast-paced action adventure. When Valve included a
better physics model into Half-Life 2, this model is neither stable enough nor
intended to be a solid representation of the real world, but most notably adds
to the fighting variety in the game.
If games are simulations, then they are social playgrounds based on given
rule sets, but even these set of rules are bound to crack under the pressure of
Chapter 1
10
the player community. Play supports behavioral studies and the bending of
the rules might produce the best results for a scientific simulation. But even
here it is difficult to relate the player’s actions back to real-world behavior.
When Blizzard added the Zul’Gurub dungeon to the virtual world of Azeroth
that is the setting for World of Warcraft (Pardo and Adham 2004), a special
fighting move of an enemy at the end of one of the new quests introduced a
virtual plague to the game world. The “virus” was transmitted over virtual
pets as well as non-player characters and player-avatars. It spread quickly
throughout the virtual world and led to the display of some real world behavior in the player community. In many ways, the resulting player behavior
in the virtual world mirrored our behavior in the real world. Players did
heal each other, for example, to contain the plague. Researchers have already
experimented with multiplayer game worlds to teach and investigate disease
control (Neulight and Kafai 2005). Yet again, direct transfer of the World
of Warcraft plague event to real-world behavior is impossible, for on the
one hand the game is limited in scope and on the other hand it differs in key
criteria of human behavior. For example, infected World of Warcraft players
could “teleport” into the middle of large cities and infect others on purpose,
making containment impossible. Whoever “died” because of the virus was
instantly reborn, which rendered the plague into an exciting in-game event
instead of a terrible tragedy—it also allowed for a different playing behavior.
Players could retaliate for being infected and the plague could turn into a
subgame in the game.
Other multiplayer worlds host growing player-driven economies that can
become highly complex. A number of problems have been caused by player
behavior that concentrates on an optimized economical development. For
example, players started “mining” in-game resources. They concentrated their
whole game efforts on gathering valuables in the game world to sell them
outside the game environment for cash. Others steal property or control over
player-avatars. Others again exploit in-game bugs. Real-world fiscal issues,
such as inflation, are also present in multiplayer game worlds. That is why
CCP Games, which runs the online world Eve Online, has hired an in-game
economist. He observes the development of the in-game economy to analyze
and steer the impact of any changes. “After we opened up an area where there
was more zydrine (an in-game mineral), we saw that price dropped. We did
not announce that there was more explicitly, but in a matter of days the price
had adjusted” (S. Hillis 2007).
All of these activities can have real financial impact as the virtual riches
often translate into goods that can be sold for real money. Through their
hard-cash value, these events have impact on the complex network of real
Introduction
11
economics. In November 2006, Ailin Graef, a prominent figure in Second Life,
who earns real money in virtual real estate deals and other in-game businesses,
announced that she had acquired more than $1 million through her online
activities (anshechung.com 2006). The result of such activity is a gradual
merging of two complex systems: the rule-driven and social networks of the
game world and economic systems of the real world. It is here that the value
of games as simulations becomes relevant, precisely because these worlds
expand beyond their rule-driven origins—because they break their original
simulation frame.
On the level of presentation and engagement, the simulation remains a
technology that often gets hacked, tweaked, and compromised by players. A
scientific simulation has to limit error margins and aim for certain neutrality
with the highest possible data density to provide valuable outcomes. In contrast, games present their world in the most appealing way to attract players,
insofar as they are simulations that dream of Hollywood, the Cirque du Soleil,
TV commercials, the Taj Mahal, or Las Vegas. 3D graphics fuel that dream
with a more and more elaborate form of presentation. While players in Tennis
for Two looked on the playing ground, modern 3D worlds take us into the
event space and onto the stage.
The Back to the Future ride was a main attraction of the Universal Studios
entertainment park in Los Angeles in the mid-1990s. The LA ride had only
opened in 1993 and combined flight simulation techniques and OMNIMAX
cinema projection in an elaborate audiovisual and haptic spectacle. Instead of
entering the airplane cockpit, visitors stepped into an oversized time-traveling
DeLorean to race through different time periods in a great spectacle directed
by Douglas Trumball and driven by Industrial Light and Magic’s effects. I
had to wait about half an hour in a queue before I could even enter the building that housed the main attraction. To sweeten the bitter pill of waiting, the
designers had multiple monitors and various movie-related paraphernalia
installed along the queue. The closer the visitor came to the main attraction,
the more one was surrounded by the fictional universe created in the Back to
the Future movies. I found myself slowly fading into the Back to the Future
universe that gradually replaced the “reality” of the theme park. When I
finally reached the last waiting line that separated the visitors into smaller
groups—each of them about to enter the pneumatic technical wonders of the
ride itself—I was already inside the Back to the Future world, thanks to the
surrounding artifacts and the expectations of the upcoming ride. Promising
rumbling, muffled music, and the screams of visitors who were just experiencing the excitement on the other side of the doors, combined to produce an
Chapter 1
12
overwhelming soundscape that was all part of the attraction. At that point,
the ground started to shake. Alarm bells added to the overall excitement.
Flashing lights went off on various corners. The doors sprang open and guides
in uniforms hurried us along short hallways. We were about to experience the
latest thrill ride on the planet—not quite.
We went right through the building and out on the other side. Only there,
blinking at the bright sunlight, along with other confused visitors, did I
slowly realize a minor earthquake had struck the park. What I had experienced
as part of the attraction was, in fact, nature’s disruption of it. The fictional
disaster ride had become the very real scene of danger (albeit minor) and reality
had played its tricks on the virtual. It had reclaimed its space.
Like game spaces, theme park attractions create an expertly directed sense
of place and context in a fictional universe within the shortest possible time.
The Back to the Future attraction did not consist solely of the main ride’s few
minutes’ duration but also of the experience of waiting, the gradual drift
into the unreal world of the attraction. Theme parks have been recognized by
the game community as important references (Carson 2000) and discussed
by a number of researchers (Darley 2000; Manovich 2001; Ndalianis 2004).
But there are fundamental differences between the two space types. The fictional space of a video game is an imagined space that lacks the physical and
nature-dependent quality of theme parks. My earthquake experience could not
have happened in the game version of the ride as presented in the GameCube
video game Universal Studios Theme Park Adventure (Okuhara 2001). The game
version of the Back to the Future attraction consists of a racing game featuring
a chase—as does the theme park version—but the game uses it merely as an
occasion to provide a whole range of other effects closer to an interactive racing
game, far removed from the physicality of the original attraction. If a real
quake would occur in the play space that houses the GameCube and my
real body, no virtual ushers would hush me along polygonal hallways into
virtual safety. The play space in the real Universal Studios and that in the
video game differ widely in the way they connect the material world and the
fictional one.
Another difference between rides and games grows out of these very real
security issues for life and safety: theme park rides restrict exploration and
interaction in order to control the balance of sensations. While players of 3D
video games can explore vast virtual landscapes and learn how to master movement in 3D, theme park rides are by and large linear experiences with no or
little interactive access. The player controls the hands and arms of the game
character and “reaches” into the game world. In contrast, visitors to theme
Introduction
13
park rides have little chance of interference with the optimized sensual spectacle they are “riding.” In the Space Mountain attraction in Disneyland Paris,
visitors literally ride the bullet with no means of interaction whatsoever. The
Space Mountain attraction offers an action-packed version of Jules Verne’s
Voyage dans la Lune that explicitly asks the visitors to “keep your arms and
hands inside the vehicle all time.” It is part of the concept that control in the
peak moments of the attraction is not in the hands of the to-be-overwhelmed
visitor but in those of the designer and engineer.
Theme park attractions that offer some form of meaningful interaction,
like DisneyQuest’s Pirates of the Caribbean: Battle for Buccaneer Gold (Fitzgerald
and Garlington 2001), turn more toward augmented reality (AR) games.
Schell and Shochet argue that these mixed forms are neither games nor rides,
but a new medium. They draw a line between theme park attraction and
interactive virtual reality (VR) pieces (Schell and Shochet 2001). Battle for
Buccaneer Gold is, in fact, an interactive piece modeled after the classic—and
far more linear—Disneyland attraction, Pirates of the Caribbean. It is not a copy
of the attraction but a more game-like in its own right installation. Like the
film version, Pirates of the Caribbean: The Curse of the Black Pearl (Verbinski
2003), which was inspired by the same attraction, the VR installation is a
different media format.
Finally, theme park attractions often use elaborate interfaces (Pausch et al.
1996; Mine 2003) that differ significantly from consumer-level video games
and their interaction design. While any attraction in a theme park is situated
in the middle of a corporate entertainment complex and visitors stream from
one waiting line to the next, most of the video games discussed here run on
home entertainment systems. That means they are played at home, in a familiar environment and with a very different interface predisposition. A theme
park ride game such as Sim Theme Park (Harris, Leinfellner, and Nuttall 2000)
includes a remediation of the ride experience, but mainly as a test of the
player’s building skills. It ends up being a reward, much like a cutscene.
Although this quotes the sensual spectacle of a real rollercoaster, it does so in
the second instance through the eye of a virtual camera. In other titles such
as the original RollerCoaster Tycoon (Sawyer 1999), the managing of the park
is the sole core of the gameplay and we see again the game-based version of
a simulation.
In order to discuss video game spaces, we need a method that allows us to
distinguish between the various spaces: one precise enough to spawn the discussion and allow for clear differentiation between game spaces and other
spaces; but also one that is scaleable enough to support more detailed critique
of a game world.
Chapter 1
14
1.3.2 Five Planes
Theorists have offered a range of approaches for analyzing video games. Each
of these approaches defines a different set of layers or planes for understanding
how games function (Konzack 2002; Montfort 2006). Any such a layering has
to be selective, because game studies are free to explore any game-related
aspect and no model can provide all possible approaches upfront. The selection
here was made with a view toward the experience of space. If “cyberspace is
a representation of human beings’ space experience” (Qvortrup 2002, 23) and
we continue such a phenomenological approach, then the system has to be
able to answer to how we perceive the space, how we are positioned in relation
to it, and the way we practice with this space. In order to address these issues,
the model suggested here distinguishes between five main conceptual planes
for the analysis of game spaces:
Figure 1.1
Five analytical planes
These planes are
1. rule-based space as defined by the mathematical rules that set, for example,
physics, sounds, AI, and game-level architecture;
Introduction
15
2. mediated space as defined by the presentation, which is the space of the
image plane and the use of this image including the cinematic form of
presentation;
3. fictional space that lives in the imagination, in other words, the space
“imagined” by players from their comprehension of the available images;
4. play space, meaning space of the play, which includes the player and the
video game hardware; and
5. social space defined by interaction with others, meaning the game space of
other players affected (e.g., in a multiplayer title).
All five are conceptual planes that have their own qualities and define themselves through different elements. Yet in order to provide a fluent gaming
experience, they all have to work in combination.
The rule-based space is defined by the code, the data, and hardware restrictions. It is the world of the functional restrictions that often mirror architectural structuring of video game spaces. This world is the basis for the mediated
space, which consists of all the output the system can provide in order to
present the rule-based game universe to the player. In the case of commercial
video games this layer consists mainly of audiovisual and tactile output that
provides a form of presentation. The player is confronted with this presentation and imagines a world from the provided information—the fictional space.
Based on the fictional world players decide on actions to affect the game space.
As long as players continue this engagement, they form a designated space in
the physical world that includes the player and the gaming system—the play
space. Finally, actions in the virtual world can affect the spaces of other players
on the layer of social space. The planes offer an analytical framework and a
perspective to approach the question of game spaces. At the same time, they
are scaleable. Each plane could be broken down into more subsections for a
finer granularity of analysis.
Others have suggested different but related categories for the understanding of space. Traces of those approaches remain in this game-space model. For
example, Lefebvre distinguishes between “spatial practice” (closest to the rulebased space), “represented space” (closest to the mediated space), and representational space (which can be read as a combination of fictional, play, and
social spaces toward an existential whole) (Lefebvre 2001). Others have chosen
to focus on specific planes. Cinematic space has been discussed, for example,
by Heath (1976) and Branigan (1992), whose findings mainly apply to the
mediated space. Huinzinga suggested the concept of the “magic circle” that
separates the world of the player from the not-playing surrounding world.
Game studies have adopted some of these philosophies (e.g., L. N. Taylor
Chapter 1
16
2002 for a reference to Lefebvre; Wolf 2002 for cinematic space; Salen and
Zimmerman 2003 for Huinzinga) and started to investigate possible relationships between separate planes (e.g., Juul 2005 for the relationship between
rule-based and fictional space). Konzack (2002) and Montfort (2006) offer
relating planes with a different focus and call for an analysis that interconnects
them. Following such a layering, the argument here will try to make connections among all five of these spaces. None of these layers alone is enough to
support a rich game world. That is why the argument will concentrate not
on a separation between these layers but on their interconnections and overlaps
to understand how they work in combination.
The two main forces that connect all layers are the two basic streams of
presentation, which leads to comprehension and motivation, and functionality, which allows for interactive access and active game worlds. Functionality
and presentation are the two forces that constantly interlink the different
planes to bring game spaces to life. Instead of focusing on any single plane,
we will look at the forces that interconnect them all. Ultimately, the division
into five principal spaces for video games provides an orientation map from
which the argument can embark. It also helps to concentrate on the given
topic of game space.
1.3.3 Cyberspace and Text Spaces
One debate that provides insight in video game space concerns cyberspace.
Discussing cyberspace within any scope is difficult. The term was derived from
fiction and never totally shed the element of the fictitious. Although there are
conceptual parallels, today’s internet does not work or look like the original
vision of cyberspace as William Gibson introduced it in 1984, yet the term
seems to work as a reference to the web and numerous other digital formats.
From the vantage point of space and spatial experience, the conflation of
Gibson’s cyberspace with the internet is questionable. Data space and navigable 3D space do not have to be the same; in fact, they rarely are. Data can
be presented and processed in many different ways: as 2D webpages, mathematical models, descriptive text, or soundscapes, for example. It was only
logical that the idea of a singular all-embracing cyberspace had to break down
into smaller and more specific segments once virtual worlds became available
en masse in video games. Once more specialized studies emerged, the idea of
a singular information space gave way to countless smaller spaces. Today there
is not one form of cyberspace.
When Dodge and Kitchin introduce their work on the mapping of
cyberspace, they point out that “at present, cyberspace does not consist of
one homogenous space; it is a myriad of rapidly expanding cyberspaces, each
Introduction
17
providing a different form of digital interaction and communication” (2001,
1). Consequently, they include maps of continuous 3D worlds such as AlphaWorlds as well as nodal networks of internet connections and website structures. Since then things have become even messier with user-generated worlds
from Counter-Strike levels to Second Life and Spore.
However, the discussion of cyberspace offers a gateway for a lot of relevant
work on virtual spaces that can serve as key references for video game spaces.
Many of these arguments remain valuable but need a readjustment to the
specifics of 3D video game worlds.
Benedikt’s seminal Cyberspace, First Steps (1991) contributes a range of
perspectives to the new phenomenon. A whole body of work on cyberspace in
the 1990s started the discussion of important issues: spiritual and theological
(e.g., Heim 1993, 1998; Wertheim 2000), architectural (e.g., Novak 1996b;
D. A. Campbell 1996), political (like Barlow’s “A Declaration of Independence of Cyberspace,” 1996), social (e.g., Rheingold 1991, 1993), cultural
(e.g., Anders 1998), theatrical (e.g., Schrum 1999), and the questions of text,
reader, and author that overlapped with some discussions in hypertext (e.g.,
Bolter 2001; Bolter and Grusin 1996). Most of these approaches will reappear
and inform the discussion.
From the earliest days of video games, the medium was driven by graphic
representation tied to rule-based processing. Titles such as Spacewar! (Graetz
and Russell 1961) or Tennis for Two—arguably among the first video games
in the sense that they could not have been realized outside the computer—
featured graphics that became classics for later clones. Their representational
form and interactive design did not derive from the literary textuality in a
literal sense, but rather from spatial realization. In Spacewar, two player-controlled spaceships try to destroy each other with torpedoes, while avoiding
collision with a central star that slowly pulls them into its gravity field. The
gameplay grows from very basic rules that are implemented in very limited
code (9 K), but it remains rich and variable for the player because these rules
are realized in a virtual space that seems to open nearly endless permutations.
For example, experienced players often used the gravity of the central star in
their opening maneuver. They let themselves be pulled toward the star while
shooting at the other spaceship. When performed by both players, the maneuver generated a shape not unlike the CBS logo—and thus was termed the
“CBS opening” (Graetz 1981).
Unlike other opening moves, for example in chess or in the initial stages
of a MUD, Spacewar simulates physical forces and an in-world timeframe that
shape the game universe. Other nondigital formats, like pen and paper roleplaying games, can try to simulate spatial behavior and map it onto their rule
Chapter 1
18
systems, but it is difficult to match the precision of the computer in the way
it handles spatial representation and functionality. Exact measurement and
orientation can be difficult variables in pen and paper games that need significant simplification, for example in the form of simplified locations on the
board. As a consequence, a lot of early video games emphasized their specific
advantage in the use of space. For example, the first mass-produced commercial arcade game Computer Space (Bushnell and Dabney 1971) brought the
simulation of the spaceflight to the foreground.
Once the game world expanded into 3D, the precision of the represented
space became the basis of many more game formats. Collisions, spatial relations of bodies to each other, audiovisual representations of the environment,
interaction with objects in the world and with the world itself—all necessarily
highlight the spatial qualities of the game. How, then, do these qualities differ
from other digital formats such as hypertext?
One way to approach video games is to look for the defining units of
meaning within them. Looking for textual units mainly in text-driven games,
Aarseth distinguishes between scriptons as “strings [of signs] as they appear to
readers” and textons as “strings as they exist in the text” (1997, 62). “A scripton, then, is an unbroken sequence of one of more textons as they are projected
by the text” (Aarseth 1994, 60). While Aarseth’s scriptons are the signs de
facto delivered to the user, he distinguishes them from the Barthesian lexia.
Barthes defines lexia as “brief, contiguous fragments . . . units of reading”
(1974, 13) that refer more to Aarseth’s textons. In reference to this division,
hypertext theorists divide hypertextual works into lexias—the individual data
segment available at a certain node—and interconnecting links (e.g., Landow
1992). The overall structure of both can result in a possible rhizome of interconnected data. The hypertext piece afternoon, a story ( Joyce 1987), for example,
consists of 539 lexia and 950 links that form a complex network of interconnections. Interactive fiction, a relative to hypertext, can generate the text and
thus might be less restrictive but still depends on descriptive textual segments
(see, e.g., Montfort 2003). These can be arranged in various ways to form
networks of lexia with different levels of complexity (Ryan 2001b). Yet none
of these levels reaches the diverse range of a 3D space.
Breaking down a 3D game space in comparable units is nearly impossible.
The options offered in a 3D navigable interactive game world include
countless opportunities, possible positions, and directions. They cannot be
separated into a definite number of textons. Rules and definitions might apply
to each element, but the combination and dynamics between objects and
worlds make such a breakdown impossible. Any movement in any direction
at any point and any specific interaction at that point change the world’s
Introduction
19
condition. Fighting games like the Tekken, Virtua Fighter, or Dead or Alive
series depend on such an optimized combination of spatial control and specific
fight moves. Any small change in a pattern might be the difference between
glorious victory and miserable failure and will trigger a different move of the
opponent. The interactive options as such are limited: spatial navigation,
blocking, and attacking moves. Their realization is unique due to their use
inside the game world and their dependency on the other character’s actions.
Instead of single nodes, game spaces put the emphasis on the forces between
innumerable states and describe them in minute detail. Elements such as
speed and direction are implemented in high detail and realized with great
effect in the game world.
For the space creation in film, Chatman distinguishes between the “literal”
story space in cinema, “that is, objects, dimensions and relations are analogous,
at least two-dimensionally, to those in the real world” (1978, 96–97) and the
“abstract” space in verbal narrative “requiring a reconstruction in the mind”
(ibid.). In this regard, games are at least as “literal” as film. Virtual set design,
lighting, visualization, and spatial sound design have become important
ingredients in the development and experience of virtual worlds. In the
context of video games, these options have changed the process of writing or
reading a text into a process of designing or exploring a space. The creator
becomes a “spacemaker” (Walser cited from Rheingold 1991, 286) or “narrative architect” (Jenkins 2004, 129) and the player an explorer and conqueror
of space. In order to identify game spaces’ specific features, the theorist as
well as the practitioner must look at their spatiality that distinguishes them
from literary pieces such as MUDs, which use spatial metaphors for orientation but consist of literary representation that offer different functionality and
presentation.
Written texts have a spatial quality but their description of space differs
significantly from the games discussed here. Moving through 3D audiovisual
video game spaces is intrinsically different from interacting with a hypertext
that uses conditional links to let the user create a path or reading through the
available node elements. When clicking on a hypertext link, the user cannot
predict the direction she will take. In fact, it is part of the pleasure of reading
(navigating) a hypertext such as Joyce’s Twilight: A Symphony (1996) to
succumb to his expertise in the writing of each node and his careful arrangement of available text elements. Clicking on any hyperlink generates a unique
reading path through the piece, but the reader’s sense of control is limited
because one cannot predict what the next move might be or where the link
might lead.
Chapter 1
20
Video game spaces are far more descriptive and graphically defined and
have developed to a stage where they contain numerous spatial references and
details. In fact, they can overwhelm us with these references. They can be so
detailed that they trap us in their perfection and their presentation can become
too flawless, too clean. They reach the uncanny valley where the distance
between nature and virtual world become reemphasized by the gradual progress toward photorealistic rendering. The first demo I saw running on the
PlayStation 2 presented a virtual forest that spread into sheer infinity—too
pixel-perfect to look alive.
Ultimately, 3D game spaces remain part of the computer as media, and
the basic textual machine that is responsible for the generation of the game
text applies to them. Game spaces and other interactive media formats rely
on the player to provide input and a system of rules that processes and reacts
to this input. But game spaces differ from other interactive media formats in
their textual units, the way they are generated by designers, and the way they
are experienced by players.
Video game spaces share qualities with simulations, theme park rides,
cyberspace, and hypertext worlds. In fact, they can be seen as simulators or a
subsection of digital space, depending on the preferences of the analyst. But
they offer different assemblies of the five defining spatial planes. These differences might be gradual—as seen in the game characteristics of the Pirates of
the Caribbean: Battle for Buccaneer Gold—but traceable. The field of video game
spaces as they are discussed here is defined by the two limitations previously
outlined: only games that are available on consumer hardware, only games
that offer navigable 3D environments. We will look at this field with the
model of the five planes as our analytical framework interconnected by presentation and functionality. The book will proceed with a discussion of these
spaces’ qualities and operational forces.
Introduction
21